Enhancing Optimal Power Flow Feasibility: A Novel Comparative Study of Approximation Techniques

Le résumé fourni par la source

This experimental study investigates three approximation methods (quadratic, linear, and logarithmic) of the Optimal Power Flow problem (OPF) in comparison to traditional DC-OPF approximations using benchmark cases from the MATPOWER library. The quality and efficiency of each method is assessed using novel error metrics in respect to the AC-OPF baseline, with a focus on phase angle, voltage magnitude and generation costs. Results demonstrate promise in outperforming DC-OPF while maintaining computational efficiency, especially in the case of the quadratic approximation. Findings indicate an attainable balance between accuracy and complexity and contribute to power system modeling with implications for real-time power system management and the integration of sustainable energy.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.